This prize is awarded annually to the company that effectively integrates analytics into organizational decision-making, and has repeatedly applied ORMS principles in pioneering, novel and lasting ways. The 2015 prize winner will describe their innovative O.R. work in a regular conference session. The 2016 winner, GM, will be recognized at the Edelman Gala on Monday evening.
Previous winners include Chevron, Memorial Sloan-Kettering Cancer Center, Sasol, Jeppesen, Intel, General Electric Global Research Center, Schneider National, Air Products and Chemicals, Procter & Gamble, UPS and other leading companies.
The George D. Smith Prize is aimed at strengthening ties between academia and industry by rewarding institutions of higher education for effective and innovative preparation of students to be good practitioners of operations research. The Prize is generously underwritten by UPS. Awarded for the first time in 2012, past winners are Sauder School of Business, University of British Columbia – Center for Operations Excellence, MIT Leaders for Global Operations, Naval Postgraduate School, and Tauber Institute for Global Operations at University of Michigan.
Three programs will compete for this year’s prize on Sunday, April 10, in Regency 5.
12:30pm Operations Research Program, United States Air Force Academy (40 min)
1:10pm Q&A with Prize Committee (15 min)
1:25pm Break & switch presenters (15 min)
1:40pm School of Information Systems & Management, and School of Public Policy & Management, H. John Heinz III College, Carnegie Mellon University (40 min)
2:20pm Q&A with Prize Committee (15 min)
2:35pm Break & switch presenters (15 min)
2:50pm Institute for Advanced Analytics, North Carolina State University (40 min)
3:30pm Q&A with Prize Committee (15 min)
3:45pm Break (15 min)
4:00pm Prize Committee Deliberations
Sponsored by CPMS, the Practice Section of INFORMS
This prize emphasizes the quality and coherence of the analysis used in practice. Dr. Wagner strove for strong mathematics applied to practical problems, supported by clear and intelligible writing. The Wagner Prize recognizes those principles by emphasizing good writing, strong analytical content and verifiable practice successes. The competition is held and the winner is announced at the INFORMS Annual Meeting in the fall. The 2015 winner, CDC, Georgia Tech, and Emory will reprise their presentation at this conference.
Past awardees include practitioners and researchers from Ford, U.S. Coast Guard, Intel, IBM T. J. Watson Research, Schneider National, Boston University, University of Florida, and others.
Sponsored by INFORMS Analytics Section
Please join us in congratulating the finalists of the 2016 Innovative Applications in Analytics Award (IAAA) sponsored by the INFORMS Analytics Section. The judging committee chaired by Scott Grasman from Rochester Institute of Technology, has selected three finalists. All finalists will be presenting their projects at the 2016 INFORMS Analytics Conference in Orlando, FL in April. Here is the list of finalists (in no particular order).
- Analytics for the Engagement Life Cycle of IBM’s Highly Valued IT Service Contracts.
Aly Megahed and Mark Smith, IBM
- Detecting Preclinical Cognitive Change.
Randall Davis and Cynthia Rudin, MIT; Dana L. Penney, Lahey Hospital and Medical Center
- Driving Organic Growth with Zilliant SalesMax.
Javier Aldrete and Lee Rehwinkel, Zilliant
The Analytics Section leadership would like to cordially thank all the members of the judging committee for their hard work in selecting these finalists.
The Innovative Applications in Analytics Award is the flagship competition of our Section. The purpose this award is to recognize the creative and unique application of a combination of analytical techniques in a new area. The prize promotes the awareness and value of the creative combination of analytics techniques in unusual applications to provide insights and business value. For more details about the competition, please visit:
Sponsored by Syngenta and the INFORMS Analytics Section
Nearly 7 million hectares of farmland are lost to soil erosion every year. Many people who produce the world’s food are living in poverty. Biodiversity is disappearing fast. And the challenge won’t get any easier: by 2050, for example, 4 billion people will be living in countries with water scarcity.
Something needs to change. We only have one planet, and we’re using its resources 50 percent faster than it can take. What we’re asking it to provide is simply not sustainable.
Each year farmers have to make decisions about what crops to plant given uncertainties in expected weather conditions and knowledge about the soil at their respective farms. These decisions have important impacts; an unusual weather pattern can have disastrous impacts on crops, but planting to hedge against stressful weather patterns can dramatically reduce yields in normal years. How can a farmer make seed variety decisions that optimally reduce risk and increase yield?
Syngenta Crop Prize Presentations will take place Monday, April 11 in Gardenia
Decision Assist Tool for Seed Variety Selection to Provide Best Yield in Known Soil and Uncertain Future Weather Conditions
Nataraju Vusirikala, Mehul Bansal, and Prathap Siva Kishore Kommi
The gap between agriculture produce and demand is ever increasing due to growing world population. There is an urgent need for utilizing all possible methods and technology solutions to bridge this gap. One of the key challenges to increase the agricultural produce is the ability to take right decisions under uncertain climate and weather conditions. In this paper we discuss a method to provide decision assist to the farmer on the best variety of soybean seed to be sown at the start of a season. In order to optimize the yield under uncertain conditions, we use a combination of crop yield modeling, weather forecasting and portfolio optimization techniques to suggest best combination of soybean seed variety. The data used in this method is the historical soybean produce data and the corresponding soil and weather conditions under which the yield was produced, day-wise weather data (temperature, precipitation and solar radiation) at farm sites from 2008 to2014. We recommend planting the following varieties with the given percentages at site 2290 for year 2016: (i) 10% of Variety V107, (ii) 35% of Variety V179, (iii) 10% of Variety V189, (iv) 10% of Variety V193, and (v) 35% of Variety V46.
Balancing Weather Risk and Crop Yield for Soybean Variety Selection
Bhupesh Shetty, Ling Tong, and Samuel Burer
We propose an optimization-based method to assist a farmer’s choice of soybean varieties to plant in order to maximize expected yield while also managing risk, where the primary uncertainty faced by the farmer is due to seasonal weather patterns. By solving a sequence of mixed-integer programs (MIPs), we calculate the efficient frontier between the two competing objectives of maximizing expected yield and maximizing the guaranteed yield over all possible season types. The coefficients of the MIPs are estimated using a multiple-linear-regression model and a Bayesian-updating scheme applied to the training and evaluation data. Using the efficient frontier, the farmer may choose an optimal solution along the frontier that fits his/her risk-reward profile.
Soy Variety Selection to Maximize Yield and Minimize Risk Based on Neural Network Prediction and Portfolio Theory
Yu Zhao, Jingsi Huang, and Ming Qin
Agricultural variety selection is vitally crucial for the markets participants in managing risks and planning operations. Here yield prediction is the most important factor and the basis for variety selection as it provides the key information for yield maximization and risk minimization. However, yield is influenced by various factors and the key factor, weather, is unpredictable for future. This paper introduces a novel model for soy variety selection. Firstly, a set of forecasting models for different varieties are developed based on the latest machine learning methods. Feature selection is used to select influence factors and three-layer feed forward neural network is constructed for yield prediction. Secondly, a portfolio theory based mixed integer programming method is proposed to find the optimal, profit-maximizing and risk-minimizing combination of soy varieties. Potentials of this model are investigated by 10 fold cross-validation. As not all the varieties are planted on the evaluation land in history, the simulation approach based on neural network forecasting model and portfolio theory provides a practical way for variety selection in reality.
The Selection of Best Soybean Varieties for Hedging Risk of Weather Uncertainties – A Deep Learning and Heuristic Optimization Approach
Mark Rees, Yidong Peng, Jeremy Babila, Mike Lyons, Lily Huang, Yinghan Song, Chun-Yang Wei, and Susan Arnot
In this study, we develop a novel approach, which integrates deep learning models and a heuristic optimization algorithm in order to decide the best portfolio of soybean varieties under the weather uncertainties. A deep learning model is learned by using historical yield data and including soybean variety, soil condition measurements, and weather condition measurements. Similarly, a second ensemble model is learned by using the transformed standard deviation of yield and the same independent variables. The two learned models are then applied to forecast the yield and standard deviation of yield at the test site for each soybean variety under different weather scenarios. Lastly, the forecasted yield and yield variance are supplied to the specially designed heuristic optimization algorithm known as glowworm swarm optimization, which selects the best combination of soybean varieties that maximize the utility function value based on the expected yield and variance of yield.
Soybean Varieties Portfolio Optimisation Based on Yield Prediction Using Weighted Histograms
Oskar Marko, Marko Panic, Sanja Brdar, and Predrag Lugonja
In this work we used a novel method for yield prediction accompanied with prediction based portfolio optimisation. For each variety we formed a predicted yield histogram, whose entries were weighted by the similarity between the evaluation and training farms. The weighted histograms method had lower error than any other method tested. The predictions were then used in portfolio optimisation. Tests showed that our method increased yield at the farm in 74.29% of cases, or for 2.86% in average. We thus recommend planting the following varieties with the given percentages: (i) 10.97% of Variety 33, (ii) 10.98% of Variety 95, (iii) 45.98% of Variety 170, (iv) 14.31% of Variety 177, and (v) 17.76% of Variety 179.
Hierarchy Modeling of Soybean Variety Yield and Decision Making for Future Planting Plan
Xiaocheng Li, Huaiyang Zhong, Stefano Ermon, and David Lobell
We introduce a novel hierarchical machine learning mechanism for predicting soybean yield that can achieve a median absolute error of 3.74 bushels per acre in five-fold cross-validation. Further, we integrate this prediction mechanism with a weather forecasting model, and propose three different approaches for decision making under uncertainty to balance yield maximization and risk. These approaches achieve similar optimal solutions over varieties V124, V41, and V44 (with a proportion of (0.2, 0.6, 0.2)), with each of which emphasizing a different aspect of the problem. In addition to efficient and accurate prediction performance, our mechanism is also highly interpretable, offering valuable insights into key properties of soybean varieties and thus a convenient way of soybean categorization.